slim-emotions-onnx

slim-emotions-onnx is a specialized function calling model that classifies the emotion of a given text context passage, and generates a python dictionary with an "emotions" key and a value of the assessed emotion, e.g., ["surprised"].

This is an ONNX int4 quantized version of slim-emotions, providing a very fast, very small inference implementation, optimized for AI PCs using Intel GPU, CPU and NPU.

Model Description

• Developed by: llmware
• Model type: tinyllama
• Parameters: 1.1 billion
• Model Parent: llmware/slim-emotions
• Language(s) (NLP): English
• License: Apache 2.0
• Uses: Emotions classifier designed for Agent-based multi-step workflows
• RAG Benchmark Accuracy Score: NA
• Quantization: int4

Model Card Contact

llmware on github

llmware on hf

llmware website